STREAM Stage 1

The evaluation of a standardised treatment regimen of anti-tuberculosis drugs for patients with MDR-TB

Can we shorten the duration of treatment and improve outcomes in patients with MDR-TB?

What was this study about?

Tuberculosis (TB) is a common, infectious, bacterial disease that is spread when an infected person transmits tubercle bacilli through the air by coughing or sneezing. Despite the availability and effectiveness of affordable six-month treatments for tuberculosis (TB), the worldwide control of this disease is currently being impacted by the emergence of multidrug resistant TB (MDR-TB).

MDR-TB refers to TB that is resistant to at least isoniazid and rifampicin. These are the two most powerful first-line drugs used to treat pulmonary TB. MDR-TB usually develops while a person is taking TB treatment due to either inappropriate treatment or failure of patients to complete their treatment.

MDR-TB is much more difficult to treat than TB strains that are sensitive to the commonly used drugs. Currently the standard treatments for MDR-TB can last as long as 24 months with a world-wide success rate of no more than about 55%. With an approximately 500,000 new cases every year there is an urgent need to develop shorter and more effective treatments.

A newly-developed 9-month treatment is being tested in this study based on a high dose fluoroquinolone and clofazimine. The aim of this investigation is to show that this shorter treatment, which has had good results in a cohort of over 500 patients in Bangladesh but has not been evaluated in a randomised trial, is at least as effective as the lengthier treatments recommended by the World Health Organisation.


What difference did this study make?

The results, based on the status of the participants at two and a half years after randomisation, were close to 80% successful in both regimens; the 9-month regimen was only 1% less effective than that of the 20 month control regimen, a difference that satisfied the predetermined criteria of non-inferiority. Participants who were coinfected with HIV did equally well on the 9 and 20 month regimen although their outcome was less good than those who were not infected with HIV. Rates of severe adverse events occurring at any time during the period of treatment or follow-up were very similar between the two regimens.

Stage 2 of STREAM is currently assessing whether a fully oral 9-month regimen, involving the new drug bedaquiline, is as effective as the 9-month regimen studied in Stage 1.

Type of study

Randomised trial

Contact details

Who funded the study?

The United States Agency for International Development (USAID) and the MRC

When did it take place?

Recruitment to STREAM Stage 1 commenced in July 2012 and was completed on 30 June 2015.

Where did it take place?

The trial took place in Ethiopia, Mongolia, South Africa and Vietnam.

Who was included?

424 people with multidrug-resistant tuberculosis were recruited.